The contribution of nitrogen fixed by annual legume pastures to the productivity of wheat in two contrasting Mediterranean environments in central Chile

Annual legume pastures exhibit high resilience and productivity in Mediterranean climate regions and can be an important source of nitrogen (N) for temperate cereals in integrated livestock production systems. This work seeks to evaluate the inputs of N fixed by annual legume mixtures (composed of 3-4 species and cultivars) grown for one, two or three years in a cropping sequence, and their contribution to soil N availability, N uptake and the grain yield of wheat, and compares this with the cereal rotation of oats and wheat with and without N fertilizer. Two field experiments comprising annual legume-wheat rotations were carried out in two contrasting Mediterranean environments in central Chile, the interior dryland in the eastern part of the coastal mountain rage, and the Andean foothills, over four growing seasons (2008-2011). At both experimental sites two mixtures of annual legumes were used in rotation with wheat. The duration of the pasture phase was between one and three years. In the interior dryland, the shoot DM of mixture M1 (T. subterraneum + M. polymorpha + T. michelianum) was higher than M2 (T. subterraneum + B. penicilius + O. compressus), and in the Andean foothills M4 (T. subterraneum + T. vesiculosum + O. compressus) was more productive than M3 (two cv. of T. subterraneum + T. incarnatum). The amount of N-2 fixed by the legume mixtures, estimated by the N-15 natural abundance technique, ranged between 28 and 87 kg N ha(-1) in the interior dryland, and between 176 and 385 kg N ha(-1) in the Andean foothills, representing respectively around 25 and 23 kg N ha(-1) fixed for every Mg of shoot DM accumulated by the annual legumes. Wheat grain yields following legume mixtures represented 53-104 % of the 3.39 Mg ha(-1) produced after oats with N fertilizer in the interior dryland, and between 68 and 101 % of 6 Mg ha(-1) attained after oats in the Andean foothills, in 2011. N use efficiency (NUE) of wheat was significantly (P <= 0.05) higher in legume-based crop rotation than after oats with or without N fertilizer, at both sites. Legume mixtures also improved the N uptake (NUPE) and N utilization (NUtE) efficiencies of wheat compared to those observed after oats with N fertilizer, at both sites. Important differences among legume mixtures and the duration of the pasture phase were observed at both sites.